DESCRIPTION: (Applicant's Abstract) We propose to continue our prospective study of HIV infection among injection drug users (IDUs) to characterize the rate of progression to immuno-suppression, AIDS, and death. Our earlier studies have examined the role of gender, age, clinical symptoms, immune activation markers, coinfections (e.g. HTLV-11), and frequency of injection drug use on the rate of developing endpoints (e.g., rapid CD4 loss, AIDS, death). We have also studied patterns of health care utilization. Here, we continue these objectives and add new foci for study including effects of type of drug used (opiates/cocaine), genetic markers (HLA haplotypes), wasting, other coinfections, and use of antiretrovirals on HIV progression. Continued follow-up is necessary to more fully describe the course of HIV infection, to permit identification of long term survivors with HIV for pathogenesis studies, to monitor temporal trends in the HIV epidemic (including the measuring of the effect of antiretroviral medications), and to increase the number of endpoints to permit focussed hypothesis testing (e.g., to identify risk factors for individual opportunistic AIDS diseases). The study population consists of 667 HIV seropositive and 209 seronegatives, of whom 330 and 140 remain alive and actively followed as of 6/95; 249 HIV seroconverters have been added between 1988-1995 (from DA05911) of whom 192 remain alive and active as of 6/95. The attendance rate at each 6 month visit exceed 90%. All participants undergo semiannual visits which include detailed behavioral interviews, medical histories, physical and gynecological examinations, and venipuncture (for laboratory assays, including T-cell subset studies, and storage of specimens in biological repository). Prospective data are summarized using survival and longitudinal data analysis; logistic regression is used to analyze nested case-control studies. We have formalized collaborations with a cohort of homosexual men in Baltimore (that are being followed with nearly identical protocols), as well as IDU cohorts in New York, Amsterdam and Italy to conduct parallel and, if appropriate, pooled analyses.